Nowadays object tracking or navigation tasks are implemented by pre-installed infrastructure like GNSS-satellites, base stations or proximity beacons or by pre-mapped environment, like radio or magnetic finger printing. However, such methods may not be efficient for indoor navigation or when sub-meter accuracy is needed.
Some applications like an auto follow camera or robotics are using image pattern recognition. There challenges are in high processing power, complex software and changing light environment and lack of capability to differentiate similar objects. Further, radio based point to point angle of arrival methods has been used over decades, like Watson-Watt method. The challenge of such methods has been in reflections and propagation distortions of radio signal. The best known implementations give good accuracy, but are still quite big, expensive and suffer from signal reflections.
Therefore, in light of the foregoing discussion, there exists a need to overcome the aforementioned drawbacks associated with tracking or navigation, particularly finding arrival direction of radio signals.